P-959. Antimicrobial Stewardship Team’s Relationship with Enterobacterales Bloodstream Infection: It's Complicated

TL;DR

This study examines how an antimicrobial stewardship team's interventions affect the treatment of uncomplicated Enterobacterales bloodstream infections.

Contribution

The study introduces a guideline for shorter antibiotic use and evaluates its impact through prospective audit and feedback.

Findings

Only 19.4% of patients with Enterobacterales bacteremia met criteria for uncomplicated infection.

Antibiotic duration was shorter for patients with infectious disease or stewardship team involvement.

Prospective audit and feedback had limited impact due to low prevalence and missed opportunities.

Abstract

Shorter antibiotic durations have shown similar clinical outcomes to longer durations for the treatment of uncomplicated Enterobacterales bloodstream infection (uE-BSI). As an antibiotic stewardship (AS) initiative, we developed an institution-approved guideline focused on identification of uE-BSI, early oral antibiotic transition, and shorter antibiotic duration with a process for prospective audit and feedback (PAF). The purpose of this study was to evaluate the incidence of uE-BSI and describe the impact of PAF at a quaternary medical center.

A real-time alert triggered by the electronic medical record (EMR) including patients with positive blood cultures for Enterobacterales was reviewed on weekdays by the AS team during 4/1/2024-9/18/2024. uE-BSI was defined as meeting all the following criteria: Known bacteremia etiology, achievement of source control, no metastatic foci, clinical improvement on appropriate antibiotic therapy for 72 hours, and immunocompetence. Patients with polymicrobial organisms mixed with non-Enterobacterales, and with situations where diagnostic work up was incomplete (e.g. death or patient directed discharge) were considered as complicated infection.

The AS team reviewed the EMR alerts for 155 patients with Enterobacterales bacteremia and 30 patients (19.4%) met criteria for uE-BSI with urinary source being the most common (70%). Among patients with uE-BSI, the AS team intervened on 6 (3.9% of all 155 patients) patients and missed 13 (8.4%) AS opportunities due to patient discharge prior to PAF. Median duration of antibiotic therapy for uE-BSI was shorter in the 17 patients who had an infectious disease (ID) consult or AS team intervention than those 13 patients who did not (8d vs. 14d, p=0.04)

At our institution, low prevalence of uE-BSI was observed and PAF did not yield robust AS interventions. However, as shorter antibiotic durations for uE-BSI were selected more frequently with ID or AS team involvement, AS initiatives to optimize the treatment of uE-BSI are beneficial. As PAF is a labor-intensive process, exploring innovative processes by leveraging the EMR to streamline the identification of patients with uE-BSI may be a useful adjunct to PAF that could increase the impact of this initiative.

All Authors: No reported disclosures